https://www.youtube.com/watch?v=5N5RGRtYxo8
video is 3 weeks old, so this isn’t outdated technically
gist: dude is 48% winrate hardstuck dia 1 - masters 4 on on account, was top 1 on all roles on another account, makes a new account, very easily wins 45 of 48 games straight into gm, while his other account is still getting impossible-to-win teams in dia 1 - masters 4.
so zero consistency in ranking/matchmaking between the accounts with insane swings (dia 1 to gm is worlds apart, and just winning 45 games in a row is statistically something that - assuming a theoretically perfectly fair matchmaking - has a mathematical chance to happen ONE in a TRILLION, lmao.
his conclusion - matchmaking is either rigged on a per-account base or just simply complete bs. This is WAY too extreme to leave room for mundane explanations aka “maybe you had not enough coffee” or similar nonsense.
1 Like
I think most people knew about this, I’ve seen countless amounts of new accounts being made and the matchmaking system is always or most of the time in their favor, you can check this sort of stuff on reddit as well as youtube, there’s plenty of proof. I’m not going to spam a load of sites etc to make my point, just look it up.
My legacy account has been on a constant loss streak, and you could argue that it’s a “skill issue” but the problem does not lie on skill, it’s quite clearly the matchmaking system as I have two other “NEWER” accounts that are miles ahead of this account. I barely play the legacy because I worry I will just be put with the wrong people, throwers, afk’ers, etc… You get it. There are too many people on here that will claim otherwise and lick Blizz’ butt hole. Clearly these type of people are getting 100% win rates.
This is the design of the matchmaker though and its working the way its intended to do, its just bad for the long standing players. The F2P model is designed so that you can keep creating new accounts and alts to reset MMR and continually get that new player bounce each season.
I wanna duplicate my reply to a video here.
Massive thx for this experiment! It’s rly confirming the state of ranking system in some way, despite some ppl would be sceptic about it. I’m back to overwatch since 7 years of not playing it. I’m not some kind of top500 player, I was ranked about plat-diamond in first seasons of Overwatch 1. When I came back I decided to immediately do the placement matches and It turns out that it was a bad decision coz I went 2 wins to 8 losses and was put into the gold rank. And since then even tho I’ve become much better at the game after bout 500 matches, I just cant get out of it. I’ve noticed that its almost no matter how good or bad I perform in a particular match, about 70% of the time the outcome will depend on the mental state of my teammates right at the start of the match. Another 30% of the time it’ll depend on how good their mechanical skills and game knowledge are, and in Gold ranks it varies from Reapers that continuously shoots the enemy into the boots to Doomfists that manages to constantly anihilate 4 of 5 ppl solo in each fight throughout the entire match (its just an examples of what I’ve witnessed). And basically it turns into a lottery at the end.
My thoughts is Overwatch ranking system need some kind of protection from that [lottery]. It must be some kind of thresholds (avg/10mins) based on collected statistic for every rank and every hero that person need to reach in a match in order to get small bonus points for a win or small discount for a loss. I know it may be bad in some cases but I think that overall it’ll be better in showing ppl that they are getting what they deserve. And it will enable a chance to climb up even with a 50/50 winrate which is the desirable rate of that system we dealing with. It might replace the “winning trend” and “losing trend” additives. Btw the “losing” one is kinda too tilting anyway when it occurs considering that you might not be that bad in all of those lost games.